Tuberculosis (TB) is the leading cause of death from a single infectious agent (Mycobacterium tuberculosis (Mtb)), causing approximately 3,000,000 deaths each year. Although TB can be effectively treated with a combination of antibiotics, drug resistant Mtb strains have recently emerged which are classified as Category C biological agents. Thus, it is widely felt that the long term control of TB will require the development of a more effective vaccine. Mycobacterium bovis Bacille Calmette-Guerin (BCG), the current anti-TB vaccine, is quite variable in its ability to protect against TB but is effective against tuberculosis meningitis, suggesting that for the foreseeable future, new TB vaccines will be given as an adjuvant or boost to BCG. Thus, understanding the immune response to both Mtb and BCG is critical for the development of an improved vaccine for TB. An increasing body of evidence indicates that both CD4+ and CD8+ T lymphocytes are critical to a protective immune response against Mtb. However, little is known about the antigens targeted by protective immune responses against Mtb in humans. Such information is required for the rational development and clinical evaluation of new, more effective TB vaccines. We propose here to characterize the human CD4+ and CD8+ T cell response to a panel of Mtb antigens in order to identify correlates with protective immunity. Antigens to be tested include proteins as well as peptide epitopes restricted by HLA-A2, an allele expressed by approximately 50% of the population. Some of these proteins and epitopes were selected from a subset of Mtb genes that are highly expressed under specified conditions and whose products are predicted to localize to the extracellular milieu, while the remainder represent previously identified HLA-A2 restricted epitopes. The T cell response to these antigens will be evaluated in peripheral blood leukocytes from three different groups of BCG immune and/or Mtb infected individuals: i. Neonates immunized a birth with one of 4 strains of BCG; ii. Individuals infected with Mtb but who do not progress to disease (latent TB infected individuals); and iii. PPD+ TB patients and PPD- "anergic" TB patients. Some of these peptide epitopes will be used to develop epitope oligomers which will be used to analyze anti-Mtb responses In vitro and in vivo. Lastly, the localization and function of Mtb peptide specific memory T cells will be studied in vivo. Correlates of protective immunity can be used to identify or prioritize protective antigens and vaccine candidates, to optimize vaccine dosing, schedules, adjuvants, etc., and to provide early evidence of efficacy. For TB, which takes years to decades to develop after infection with Mtb, immune correlates with protection are an attractive, and perhaps essential, supplement to efficacy trials.